Video Caption MCP Server

Generate captions and subtitles from user-uploaded audio/video files. Translate existing captions to other languages. Powered by Whisper (MIT) + Qwen (Apache 2.0).

IMPORTANT — Legal Constraint

❌ NO downloads from YouTube, Vimeo, TikTok, Twitch, or any video platform.

• Only user-hosted audio/video files are accepted (your own server, S3, etc.)
• All platform URLs are automatically rejected
• You must own or have permission to caption the content
• Captions are AI-generated and may contain errors

Features

Tool	Description	Price
generate_captions	Generate SRT/VTT captions from audio/video	$0.08/use
translate_captions	Translate existing SRT/VTT to another language	$0.10/use

Connect via Claude Desktop

Add to your Claude Desktop MCP settings:

{
  "mcpServers": {
    "video-captions": {
      "url": "https://ntriqpro--video-caption-mcp.apify.actor/mcp?token=YOUR_APIFY_TOKEN"
    }
  }
}

Supported Audio/Video Formats

MP3, WAV, M4A, OGG, FLAC, WebM, MP4, MOV, AVI, MKV, and other common formats.

Tool Usage

generate_captions

Generates SRT or VTT subtitle file from audio/video.

Input:

• audio_url (required): URL to your hosted audio/video file (not platform URLs)
• format (optional): "srt" or "vtt" (default: "srt")

Output Example (SRT):

1
00:00:00,000 --> 00:00:03,500
Hello, welcome to today's meeting.

2
00:00:03,500 --> 00:00:07,200
Let's start with the agenda items.

Output Example (VTT):

WEBVTT

00:00:00.000 --> 00:00:03.500
Hello, welcome to today's meeting.

00:00:03.500 --> 00:00:07.200
Let's start with the agenda items.

Full Response:

{
  "status": "success",
  "subtitles": "1\n00:00:00,000 --> 00:00:03,500\nHello, welcome...",
  "format": "srt",
  "segment_count": 15,
  "language": "en",
  "model": "whisper + qwen",
  "disclaimer": "You must own or have permission to caption the provided content. We do not download from video platforms. Captions are AI-generated and may contain errors."
}

translate_captions

Translates SRT/VTT captions to another language.

Input:

• subtitle_content (required): Full SRT or VTT text
• target_language (required): Language code (e.g., "es", "fr", "de", "ja", "ko", "zh", "pt", "ru")

Full Response:

{
  "status": "success",
  "translated": "1\n00:00:00,000 --> 00:00:03,500\nHola, bienvenido a la reunión de hoy...",
  "target_language": "es",
  "source_language": "en",
  "model": "qwen",
  "disclaimer": "You must own or have permission to caption the provided content. We do not download from video platforms. Captions are AI-generated and may contain errors."
}

File Upload Methods

Option 1: Direct HTTP URL

https://example.com/my-audio.mp3
https://my-bucket.s3.amazonaws.com/video.mp4

Option 2: AWS S3 Pre-Signed URL

https://my-bucket.s3.amazonaws.com/video.mp4?X-Amz-Algorithm=AWS4-HMAC-SHA256&...

Option 3: Google Cloud Storage Public URL

https://storage.googleapis.com/my-bucket/audio.mp3

Error Handling

Platform URL Blocked

{
  "error": "Platform URLs not supported. Please upload your own audio/video file."
}

Invalid URL

{
  "error": "Invalid audio_url. Must be a valid HTTP/HTTPS URL."
}

Empty Subtitle Content

{
  "error": "subtitle_content cannot be empty."
}

Technology Stack

• Speech Recognition: OpenAI Whisper (MIT License)
• Text Translation: Qwen 3.5 (Apache 2.0 License)
• Processing: Cloud-based inference via ai.ntriq.co.kr
• Privacy: Files processed in real-time, not stored

Open Source Licenses

This service uses the following open source models:

• OpenAI Whisper — MIT License
• Qwen 3.5 — Apache 2.0 License

Legal & Disclaimer

Ownership & Consent: Users are solely responsible for ensuring:

• They own or have permission to caption the provided content
• Compliance with all applicable laws regarding content usage
• No copyrighted material is submitted without authorization

AI-Generated Content: Captions and translations are AI-generated and may:

• Contain errors or inaccuracies
• Misinterpret context or idioms
• Require manual review before publishing

Data Processing:

• Audio/video files are processed in real-time
• Files are not stored or retained
• Files are not used for model training
• This service is a tool only and does not constitute legal or compliance advice

Platform Constraint:

• This tool explicitly does NOT support downloading or processing videos from YouTube, Vimeo, TikTok, Twitch, or similar platforms
• Use only with content you own or have explicit permission to process
• Violating platform Terms of Service is the user's responsibility

Pricing

• generate_captions: $0.08 per use
• translate_captions: $0.10 per use
• Usage is Pay-Per-Event (PPE) — only charged when tools are called